The professional sports and television industry place an increased emphasis on enhanced viewer experiences to retain and grow their audiences thus growing their advertising revenue. Also, time shifted television viewing, in which advertisements are often skipped, is increasing and the viewer's willingness to sit through a three hour sporting event is declining. Advertisements used in conjunction with enhanced viewer experience may help counter this growing trend. Many ways of enhanced viewer experiences in sporting events are in practice today. The most popular may be the First Down Line projected on the football field image or the superimposition of two downhill skiing runs to provide the excitement of virtual racing in a non-simultaneous event.
Arguably, the most popular televised sporting event in the United States is professional football. It is certainly one in which a tremendous amount of money and effort is placed on enhanced viewer experience.
FIG. 1 illustrates an example frame 100 of a live football broadcast using a high angle view from the sidelines. Frame 100 includes football player 102, a football field 104, the first down marker 106 and First Down Line 108. First Down Line 108 is a computer generated image superimposed on football field 104 representing an estimated location of the first down marker 106 in relation to the whole football field 104.
Providing the ability to see the football player 102 in relation to First Down Line 108 while running up the field is an extremely effective way to enhance viewer experience. The process of placing First Down Line 108 is far from simple. On game day, it requires a minimum of four people and a plurality of computers to produce the desired effect. First Down Line 108 can be even more difficult if snow or mud is involved, because the technique relies upon color differences to distinguish the players 102 from the green field 104.
FIG. 2 illustrates an example frame 200 of a frontal view of a live football broadcast. Frame 200 includes a football player 202 and a football field 204. The small square 206 on a portion of football 208 represents one picture element (pixel) which will be discussed in more detail later.
The frontal view represented by frame 200 in FIG. 2 offers an exciting view of the football game and is a conventional image used often during sporting broadcasts. A problem that occurs with the prior art system in FIG. 1 is that the frontal view is not particularly compatible with First Down Line 108.
Another problem with the conventional frontal view illustrated in FIG. 2 is that the viewer has limited depth perception in the two-dimensional image of the three-dimensional scene. Accordingly, even if First Down Line 108 is incorporated into football field 204, it would be difficult for the viewer to know the location of football player 202 in reference to First Down Line 108 or other objects in frame 200.
FIG. 3 illustrates a projected pixel of an image projected onto a single sensor within a video camera. In the figure, a single sensor 304 captures a projected image 302. The direction of projected image 302 is represented by line 310. However, the depth of the projected image 302 cannot be accurately calculated with single sensor 304. Projected image 302 could correspond to one of an infinite number of imaged objects. For example, projected image 302 may correspond to a first object 306 or may correspond to a second object 308. First object 306 is closer to single sensor 304 than second object 308. Further, first object 306 is smaller than second object 308. In such a case, first object 306 may be projected onto sensor 304 as the same size as a projected image of second object 308.
Assume projected image 302 corresponds to pixel 206 on portion of football 208. The sum of all pixels on football player 202 would become the entire image of football player 202 as shown in frame 200. Frame 200 is produced from a single sensor and therefore it is difficult to tell exactly where the football player 202 is in relation to the other objects and football players on frame 200. There is no depth perception.
What is needed is a way to convey to the viewer a perception of depth in a two dimensional image.